From a7f3f939dd7d8398acebecd1ceb2e9e7ffbe91d2 Mon Sep 17 00:00:00 2001 From: Ray Jui Date: Fri, 1 Dec 2017 03:13:02 -0500 Subject: [PATCH 1/4] bnxt_en: Need to unconditionally shut down RoCE in bnxt_shutdown The current 'bnxt_shutdown' implementation only invokes 'bnxt_ulp_shutdown' to shut down RoCE in the case when the system is in the path of power off (SYSTEM_POWER_OFF). While this may work in most cases, it does not work in the smart NIC case, when Linux 'reboot' command is initiated from the Linux that runs on the ARM cores of the NIC card. In this particular case, Linux 'reboot' results in a system 'L3' level reset where the entire ARM and associated subsystems are being reset, but at the same time, Nitro core is being kept in sane state (to allow external PCIe connected servers to continue to work). Without properly shutting down RoCE and freeing all associated resources, it results in the ARM core to hang immediately after the 'reboot' By always invoking 'bnxt_ulp_shutdown' in 'bnxt_shutdown', it fixes the above issue Fixes: 0efd2fc65c92 ("bnxt_en: Add a callback to inform RDMA driver during PCI shutdown.") Signed-off-by: Ray Jui Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index c5c38d4b7d1c..7f173eb42aa2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -8263,8 +8263,9 @@ static void bnxt_shutdown(struct pci_dev *pdev) if (netif_running(dev)) dev_close(dev); + bnxt_ulp_shutdown(bp); + if (system_state == SYSTEM_POWER_OFF) { - bnxt_ulp_shutdown(bp); bnxt_clear_int_mode(bp); pci_wake_from_d3(pdev, bp->wol); pci_set_power_state(pdev, PCI_D3hot); From c8fb7b8259c67b86cd93a71c85e78b34d2c96fdc Mon Sep 17 00:00:00 2001 From: Sunil Challa Date: Fri, 1 Dec 2017 03:13:03 -0500 Subject: [PATCH 2/4] bnxt_en: wildcard smac while creating tunnel decap filter While creating a decap filter the tunnel smac need not (and must not) be specified as we cannot ascertain the neighbor in the recv path. 'ttl' match is also not needed for the decap filter and must be wild-carded. Fixes: f484f6782e01 ("bnxt_en: add hwrm FW cmds for cfa_encap_record and decap_filter") Signed-off-by: Sunil Challa Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index d5031f436f83..96bff48af971 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -532,10 +532,8 @@ static int hwrm_cfa_decap_filter_alloc(struct bnxt *bp, } if (flow->flags & BNXT_TC_FLOW_FLAGS_TUNL_ETH_ADDRS) { - enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_DST_MACADDR | - CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_SRC_MACADDR; + enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_DST_MACADDR; ether_addr_copy(req.dst_macaddr, l2_info->dmac); - ether_addr_copy(req.src_macaddr, l2_info->smac); } if (l2_info->num_vlans) { enables |= CFA_DECAP_FILTER_ALLOC_REQ_ENABLES_T_IVLAN_VID; @@ -1012,10 +1010,9 @@ static int bnxt_tc_get_decap_handle(struct bnxt *bp, struct bnxt_tc_flow *flow, if (rc) goto put_decap; - decap_key->ttl = tun_key.ttl; decap_l2_info = &decap_node->l2_info; + /* decap smac is wildcarded */ ether_addr_copy(decap_l2_info->dmac, l2_info.smac); - ether_addr_copy(decap_l2_info->smac, l2_info.dmac); if (l2_info.num_vlans) { decap_l2_info->num_vlans = l2_info.num_vlans; decap_l2_info->inner_vlan_tpid = l2_info.inner_vlan_tpid; From e9ecc731a87912d209d6e9b4ed20ed70451c08cb Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Fri, 1 Dec 2017 03:13:04 -0500 Subject: [PATCH 3/4] bnxt_en: fix dst/src fid for vxlan encap/decap actions For flows that involve a vxlan encap action, the vxlan sock interface may be specified as the outgoing interface. The driver must resolve the outgoing PF interface used by this socket and use the dst_fid of the PF in the hwrm_cfa_encap_record_alloc cmd. Similarily for flows that have a vxlan decap action, the fid of the incoming PF interface must be used as the src_fid in the hwrm_cfa_decap_filter_alloc cmd. Fixes: 8c95f773b4a3 ("bnxt_en: add support for Flower based vxlan encap/decap offload") Signed-off-by: Sathya Perla Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c | 48 +++++++++++--------- 1 file changed, 26 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index 96bff48af971..3d201d7324bd 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -56,7 +56,6 @@ static int bnxt_tc_parse_redir(struct bnxt *bp, { int ifindex = tcf_mirred_ifindex(tc_act); struct net_device *dev; - u16 dst_fid; dev = __dev_get_by_index(dev_net(bp->dev), ifindex); if (!dev) { @@ -64,15 +63,7 @@ static int bnxt_tc_parse_redir(struct bnxt *bp, return -EINVAL; } - /* find the FID from dev */ - dst_fid = bnxt_flow_get_dst_fid(bp, dev); - if (dst_fid == BNXT_FID_INVALID) { - netdev_info(bp->dev, "can't get fid for ifindex=%d", ifindex); - return -EINVAL; - } - actions->flags |= BNXT_TC_ACTION_FLAG_FWD; - actions->dst_fid = dst_fid; actions->dst_dev = dev; return 0; } @@ -160,13 +151,17 @@ static int bnxt_tc_parse_actions(struct bnxt *bp, if (rc) return rc; - /* Tunnel encap/decap action must be accompanied by a redirect action */ - if ((actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP || - actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP) && - !(actions->flags & BNXT_TC_ACTION_FLAG_FWD)) { - netdev_info(bp->dev, - "error: no redir action along with encap/decap"); - return -EINVAL; + if (actions->flags & BNXT_TC_ACTION_FLAG_FWD) { + if (actions->flags & BNXT_TC_ACTION_FLAG_TUNNEL_ENCAP) { + /* dst_fid is PF's fid */ + actions->dst_fid = bp->pf.fw_fid; + } else { + /* find the FID from dst_dev */ + actions->dst_fid = + bnxt_flow_get_dst_fid(bp, actions->dst_dev); + if (actions->dst_fid == BNXT_FID_INVALID) + return -EINVAL; + } } return rc; @@ -899,10 +894,10 @@ static void bnxt_tc_put_decap_handle(struct bnxt *bp, static int bnxt_tc_resolve_tunnel_hdrs(struct bnxt *bp, struct ip_tunnel_key *tun_key, - struct bnxt_tc_l2_key *l2_info, - struct net_device *real_dst_dev) + struct bnxt_tc_l2_key *l2_info) { #ifdef CONFIG_INET + struct net_device *real_dst_dev = bp->dev; struct flowi4 flow = { {0} }; struct net_device *dst_dev; struct neighbour *nbr; @@ -1006,7 +1001,7 @@ static int bnxt_tc_get_decap_handle(struct bnxt *bp, struct bnxt_tc_flow *flow, */ tun_key.u.ipv4.dst = flow->tun_key.u.ipv4.src; tun_key.tp_dst = flow->tun_key.tp_dst; - rc = bnxt_tc_resolve_tunnel_hdrs(bp, &tun_key, &l2_info, bp->dev); + rc = bnxt_tc_resolve_tunnel_hdrs(bp, &tun_key, &l2_info); if (rc) goto put_decap; @@ -1092,8 +1087,7 @@ static int bnxt_tc_get_encap_handle(struct bnxt *bp, struct bnxt_tc_flow *flow, if (encap_node->tunnel_handle != INVALID_TUNNEL_HANDLE) goto done; - rc = bnxt_tc_resolve_tunnel_hdrs(bp, encap_key, &encap_node->l2_info, - flow->actions.dst_dev); + rc = bnxt_tc_resolve_tunnel_hdrs(bp, encap_key, &encap_node->l2_info); if (rc) goto put_encap; @@ -1166,6 +1160,15 @@ static int __bnxt_tc_del_flow(struct bnxt *bp, return 0; } +static void bnxt_tc_set_src_fid(struct bnxt *bp, struct bnxt_tc_flow *flow, + u16 src_fid) +{ + if (flow->actions.flags & BNXT_TC_ACTION_FLAG_TUNNEL_DECAP) + flow->src_fid = bp->pf.fw_fid; + else + flow->src_fid = src_fid; +} + /* Add a new flow or replace an existing flow. * Notes on locking: * There are essentially two critical sections here. @@ -1201,7 +1204,8 @@ static int bnxt_tc_add_flow(struct bnxt *bp, u16 src_fid, rc = bnxt_tc_parse_flow(bp, tc_flow_cmd, flow); if (rc) goto free_node; - flow->src_fid = src_fid; + + bnxt_tc_set_src_fid(bp, flow, src_fid); if (!bnxt_tc_can_offload(bp, flow)) { rc = -ENOSPC; From ebd5818cc5d4847897d7fe872e2d9799d7b7fcbb Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Fri, 1 Dec 2017 03:13:05 -0500 Subject: [PATCH 4/4] bnxt_en: Fix a variable scoping in bnxt_hwrm_do_send_msg() short_input variable is assigned to another data pointer which is referred out of its scope. Fix it by moving short_input definition to the beginning of bnxt_hwrm_do_send_msg() function. No failure has been reported so far due to this issue. Fixes: e605db801bde ("bnxt_en: Support for Short Firmware Message") Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 7f173eb42aa2..28f5e94274ee 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3368,6 +3368,7 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, u16 cp_ring_id, len = 0; struct hwrm_err_output *resp = bp->hwrm_cmd_resp_addr; u16 max_req_len = BNXT_HWRM_MAX_REQ_LEN; + struct hwrm_short_input short_input = {0}; req->seq_id = cpu_to_le16(bp->hwrm_cmd_seq++); memset(resp, 0, PAGE_SIZE); @@ -3376,7 +3377,6 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, if (bp->flags & BNXT_FLAG_SHORT_CMD) { void *short_cmd_req = bp->hwrm_short_cmd_req_addr; - struct hwrm_short_input short_input = {0}; memcpy(short_cmd_req, req, msg_len); memset(short_cmd_req + msg_len, 0, BNXT_HWRM_MAX_REQ_LEN -